Dr. Bimal Jalan - Governor, Reserve Bank of India     


Exchange Rate Management: An Emerging Consensus?*

I am very happy to be here with you on the occasion of the 14th National Assembly. This annual gathering of our Forex specialists is a very special occasion for everyone interested in the evolution of forex market in India – not only the dealers and market participants, but also the RBI, Government and outside experts interested in appropriate macro-economic management for higher growth and the greater good of our people. It provides all of us with an opportunity to review recent developments in forex markets against the background of global developments in a fast changing world economic scenario and modify our own policy and approach. I am, therefore, delighted to be with you once again.

2. As you know, RBI is in regular touch, formally and informally, with your association to review market developments. My senior colleagues and I have also benefited from your advice and, as you are aware, a large number of measures have now been put in place to liberalise our regulations in respect of foreign exchange transactions. With your permission, today, on this special occasion, I would like to deal briefly with some of the longer term policy issues in respect of exchange rate and reserves management. Several of these issues have also been debated in various international fora. In the context of upsurge in our reserves in recent years and the appreciating trend in the external value of the rupee, there has also been considerable domestic discussion of these issues.

3. You will recall that the last time I had addressed your Association was in December 2000 - nearly 2 ˝ years ago – on the occasion of the 21st Asia Pacific Forex Congress. That meeting was taking place against the background of the Asian crisis in 1997-98. The Asian region had just come out of the forex crisis of a very destabilising kind. I had used that occasion to review the on-going debate on management of the external sector, particularly the appropriate exchange rate systems, the appropriate intervention policy, and the foreign exchange reserve policy. Soon after the Asian crisis, these subjects had figured very prominently in the discussions on International Financial Architecture in various fora, such as IMF, the World bank, G20, Financial Stability Forum and the Bank for International Settlements.

4. At that time, while reviewing the state of the debate, I had mentioned that a worldwide consensus on several issues was still evolving. Today, while "consensus" may be too strong a word, I believe that there is a fair degree of convergence in the dominant international opinion among experts and various specialised institutions on many of these issues. Let me summarise some of the main conclusions which have emerged in the last 2 or 3 years on the management of the external sector. Once again, I should emphasise that, given widely different economies of the world that we are talking about, there is no global consensus as such or unity of views. However, at present, as I see it, there is certainly a "dominant" view of what is right and appropriate, which is increasingly commanding international acceptance.

5. Thus, on the question of the appropriate exchange rate regime, a fixed exchange rate regime (even with a Currency Board) is clearly out of favour. The Brazilian and Argentinian crisis, after the Asian crisis, came as a rude shock. Even strong Currency Board type arrangements of a fixed peg vis-à-vis dollar were found to be unviable. You will recall that, soon after the Asian crisis, the widely accepted theoretical position was that a country had the choice of either giving up monetary independence and setting up a Currency Board or giving up the stable currency objective and letting the exchange rate float freely so that monetary policy could then be directed to the objectives of inflation control. There is a shift in this paradigm. The possibility of having a viable fixed rate mechanism has been generally discarded, and the dominant view now is that, for most countries floating or flexible rates are the only sustainable way of having a less crisis-prone exchange rate regime.

6. In regard to the desirable degree of flexibility in exchange rates, opinions and practices vary. But a completely "free" float, without intervention, is clearly out of favour except perhaps in respect of a few global or reserve currencies. And, in respect of these currencies also (say, Euro and Dollar), concerns are expressed at the highest levels if the movement is sharp in either direction – recently, for example, when Euro was strengthening at a fast pace. Studies by the IMF and several experts also show that by far, the most common exchange rate regime adopted by countries, including industrial countries, is not a free float. Most of the countries have adopted intermediate regimes of various types, such as, managed floats with no pre-announced path, and independent floats with foreign exchange intervention moderating the rate of change and preventing undue fluctuations. By and large, barring a few, countries have "managed" floats and Central Banks intervene periodically. This has also been true of industrial countries. In the past, the U.S., the EU and the U.K. have also intervened at one time or another. Thus, irrespective of the pure theoretical position in favour of a free float, the external value of the currency continues to be a matter of concern to most countries, and most central banks.

7. The reason why intervention by most central banks in forex markets has become necessary from time to time is primarily because of two reasons. A fundamental change that has taken place in recent years is the importance of capital flows in determining exchange rate movements as against trade deficits and economic growth, which were important in the earlier days. The latter do matter, but only over a period of time. Capital flows, on the other hand, have become the primary determinants of exchange rate movements on a day-to-day basis. Secondly, unlike trade flows, capital flows in "gross" terms which affect exchange rate can be several times higher than "net" flows on any day. These are also much more sensitive to what everybody else is saying or doing than is the case with foreign trade or economic growth. Therefore, herding becomes unavoidable. I am sure you will agree that all dealers prefer to be wrong with everyone else rather than being wrong alone!

8. A related issue, which is a corollary of the prevalent intermediate regimes in respect of exchange rates, concerns the need, if any, for foreign exchange reserves. In a regime of free float, it could be argued that there was really no need for reserves. If demand for foreign exchange is higher than supply, exchange rates will depreciate and equilibrate demand and supply over time. If supply exceeded demand, exchange rates will appreciate and sooner or later, the two will equalise at some price. However, today in the light of volatility induced by capital flows and the self-fulfilling expectations that this can generate, there is now a growing consensus that emerging market countries should, as a matter of policy, maintain "adequate" reserves. How adequacy is to be defined is also becoming clearer. Earlier, the rule used to be defined in terms of number of months of imports. Now, increasingly it is felt that reserves should at least be sufficient to cover likely variations in capital flows or the "liquidity-at-risk". (However, there is as yet no consensus on the upper limit for reserves. Even after an "adequate" level is reached, reserves may continue to increase if capital inflows are strong and central banks decide to intervene in order to moderate the degree of appreciation.)

9. To sum up, it seems that the debate on appropriate policies relating to forex markets has now converged around some generally accepted views. Among these, as I mentioned are: (a) exchange rates should be flexible and not fixed or pegged; (b) countries should be able to intervene or manage exchange rates – to at least some degree – if movements are believed to be destablising in the short run; and (c) reserves should at least be sufficient to take care of fluctuations in capital flows and "liquidity at risk".

10. Let me now briefly deal with some issues of practical importance in the management of forex markets in India, which have figured prominently in the media and expert commentary. I will first take up 2 or 3 matters which are rather straight-forward and on which our policy stance is also equally "unambiguous" and clear-cut. Then I will move to, and conclude with, a discussion of the appropriate exchange rate policy for India in the current situation.

11. A frequently discussed question is about Capital Account Convertibility (CAS), i.e. when is India going to move to full CAC? As you are aware, we have already liberalized and deregulated a whole host of capital account transactions. It is probably fair to say that for most transactions which are required for business or personal convenience, the rupee is, for all practical purposes, convertible. In cases, where specific permission is required for transactions above a high monetary ceiling, this permission is also generally forthcoming. It is also the declared policy of the Government and the RBI to continue with this process of liberalization. In this sense, Capital Account Convertibility continues to be a desirable objective for all investment and business related transactions and India should be able to achieve this objective in not too distant a future.

12. There are, however, two areas where we would need to be extremely cautious – one is unlimited access to short-term external commercial borrowing for meeting working capital and other domestic requirements. The other area concerns the question of providing unrestricted freedom to domestic residents to convert their domestic bank deposits and idle assets (such as, real estate), in response to market developments or exchange rate expectations.

13. In respect of short-term external commercial borrowings, there is already a strong international consensus that emerging markets should keep such borrowings relatively small in relation to their total external debt or reserves. Many of the financial crises in the 1990s occurred because the short-term debt was excessive. When times were good, such debt was easily accessible. The position, however, changed dramatically in times of external pressure. All creditors who could redeem the debt did so within a very short period, causing extreme domestic financial vulnerability. The occurrence of such a possibility has to be avoided, and we would do well to continue with our policy of keeping access to short-term debt limited as a conscious policy at all times – good and bad.

14. So far as the free convertibility of domestic assets by residents is concerned, the issues are somewhat more fundamental. It has to do with the differential impact of "stock" and "flows" in determining external vulnerability. The day-to-day movement in exchange rates is determined by "flows" of funds, i.e. by demand and supply of spot or forward transactions in the market. Now, suppose the exchange rate is depreciating unduly sharply (for whatever reasons) and is expected to continue to do so for the near future. Now, further suppose that domestic residents, therefore, decide – perfectly rationally and reasonably – that they should convert a part or whole of their stock of domestic assets from domestic currency to foreign currency. This will be financially desirable as the domestic value of their converted assets is expected to increase because of anticipated depreciation. And, if a large number of residents so decide simultaneously within a short period of time, as they may, this expectation would become self-fulfilling. A severe external crisis is then unavoidable.

15. Consider India’s case, for example. Today, our reserves are high and exchange rate movements are, by and large, orderly. Now, suppose there is an event which creates external uncertainty, as for example, what actually happened at the time of the Kargil or the imposition of sanctions after Pokhran, or the oil crises earlier. Domestic stock of bank deposits in rupees in India is presently close to US $ 290 billion, nearly three and a half times our total reserves. At the time of Kargil or Pokhran or the oil crises, the multiple of domestic deposits over reserves was in fact several times higher than now. One can imagine what would have had happened to our external situation, if within a very short period, domestic residents decided to rush to their neighbourhood banks and convert a significant part of these deposits into sterling, euro or dollar.

16. No emerging market exchange rate system can cope with this kind of contingency. This may be an unlikely possibility today, but it must be factored in while deciding on a long term policy of free convertibility of "stock" of domestic assets. Incidentally, this kind of eventuality is less likely to occur in respect of industrial countries with international currencies such as Euro or Dollar, which are held by banks, corporates, and other entities as part of their long-term global asset portfolio (as distinguished from emerging market currencies in which banks and other intermediaries normally take a daily long or short position for purposes of currency trade).

17. Another issue, which has figured prominently in the current debate, relates to foreign exchange reserves. As is well known, India’s foreign exchange reserves have increased substantially in the past few years and are now among one of the largest in the world. The fact that most of the constituents of India’s balance of payments are showing positive trends – on the current as well as capital accounts – is a reflection of the increasing competitiveness of the Indian economy and strong confidence of the international community in India’s growth potential. For the first time after our Independence 56 years ago, the fragility of the balance of payments is no longer a concern of policy makers. This is a highly positive development and regarded as such by the country at large.

18. Nevertheless, there are two concerns that have been expressed by expert commentators – one is about the "cost" of additional reserves, and second concerns the impact of "arbitrage" in inducing higher inflows. So far as the cost of additional reserves is concerned, it needs to be borne in mind that the bulk of additions to reserves in the recent period is on account of non-debt creating inflows. India’s total external debt, including NRI (Non-Resident Indian) deposits, has increased relatively slowly as compared with the increase in reserves, particularly in the last couple of years. In fact, India pre-paid more than $ 3 billion of external debt earlier this year. It may also be mentioned that rates of interest paid on NRI deposits and multilateral loans in foreign currency are in line with or lower than prevailing international interest rates.

19. On NRI rupee deposits, interest rates in the last couple of years have been in line with interest rates on deposits by residents, and are currently even lower than domestic interest rates. So far as other non-debt creating inflows (i.e., foreign direct investment, portfolio investment or remittances) are concerned, such inflows by their very nature are commercial in nature and enjoy the same returns and risks, including exchange rate risk, as any other form of domestic investment or remittance by residents. The cost to the country of such flows is the same whether they are added to reserves or are matched by equivalent foreign currency outflow on account of higher imports or investments abroad by residents. On the whole, under present conditions, it seems that the "cost" of additional reserves is really a non-issue from a broader macro-economic point of view.

20. Indian interest rates have come down substantially in the last three or four years. They are, however, still higher than those prevailing in the U.S., Europe, U.K. or Japan. This provides an "arbitrage" opportunity to holder of liquid assets abroad, who may take advantage of higher domestic interest rates in India leading to a possible short-term upsurge in capital flows. However, there are several considerations, which indicate that "arbitrage" per se is unlikely to have been a primary factor in influencing remittances or investment decisions by NRIs or foreign entities in the recent period. Among these are :

  • The minimum period of deposits by NRIs in Indian rupees is now one year, and the interest rate on such deposits is subject to a ceiling rate of 2.5 per cent over Libor. This is broadly in line with one-year forward premium on the dollar in the Indian market (interest rates on dollar deposits by NRIs are actually below Libor).

  • Outside of NRI deposits, investments by Foreign Institutional Investors (FIIs) in debt funds is subject to an overall cap of only $ 1 billion in the aggregate. In other words, the possibility of arbitrage by FIIs in respect of pure debt funds is limited to this low figure of $ 1 billion (excluding investments in a mix of equity and debt funds).

  • Interest rates and yields on liquid securities are highly variable abroad as well as in India, and the differential between the two rates can change very sharply within a short time depending on market expectations. It is interesting to note that the yield on 10 year Treasury bills in the U.S. had risen to about 4.4 per cent as compared with 5.6 per cent on Government bonds of similar maturity in India at the end of July 2003. Taking into account the forward premia on dollars and yield fluctuations, except for brief period, there is likely to be little incentive to send large amounts of capital to India merely to take advantage of the interest differential.

21. On the whole, it is likely that external flows into India have been motivated by factors other than pure arbitrage. Figures on sources of reserve accretion available upto the end of last year (2002-03) confirm this view. It is also pertinent to note that domestic interest rates among industrial countries also vary considerably. For example, in Japan, they are close to zero. In the U.K., they are above 4 per cent, and in the U.S. about 1.5 per cent. There is no evidence that capital has been moving out of U.S. to U.K. or Europe merely on account of interest differential. Within a certain low range, capital flows are likely to be more influenced by outlook for growth and inflation than pure arbitrage even among industrial countries with full CAC.

22. Another point which has been forcefully put forward by several experts in the context of rising reserves, is that India should use its reserves for increasing investment for further development of the country rather than keep them as liquid assets. It is argued that it is paradoxical for a developing country to have a current and capital account surplus, and thereby add to its reserves, rather than use foreign savings to enhance the rate of investment in the economy.

23. In principle, this point is valid. There is no doubt that in our present situation, maximum support has to be given to increasing the level of investment, particularly in the infrastructure sector. It is for this reason that RBI in the recent period has been following a soft interest rate policy in an environment of low inflation. However, at the same time, it must be emphasized that there is very little that RBI, (or, for that matter, Government) can directly do to use additional reserves for investment. The equivalent rupee resources have already been released by the RBI to recipients of foreign exchange, and equivalent rupee liquidity has already been created. The decision on whether to invest, consume or deposit these additional rupee resources lies with recipients, and not with the RBI. By all means, let us urge them to invest, but there is not much of a case for pointing a finger at additional reserves as a "cause" of lower than desirable level of investment activity in the economy.

24. Let me now come to my last point, which is of considerable present-day interest in India in the context of high and rising reserves, easy liquidity, low interest rates and the weakening dollar, i.e., what should be the correct or right policy stance for the management of exchange rate in India in the present environment? In RBI’s periodic credit policy statements, as well as other public statements, RBI has highlighted the main pillars of its strategy for the management of the exchange rate. These are: RBI does not have a fixed "target" for the exchange rate which it tries to defend or pursue over time; RBI is prepared to intervene in the market to dampen excessive volatility as and when necessary; RBI’s purchases or sales of foreign currency are undertaken through a number of banks and are generally discreete and smooth; and market operations and exchange rate movement should, in principle, be transaction-oriented rather than purely speculative in nature.

25. It is perhaps fair to say that the actual results of the exchange rate policy followed by the RBI, since the Asian crisis in particular, have been highly positive so far. In addition to sharp increase in reserves and generally "orderly" movements in exchange rates with lower volatility, the confidence level of domestic and foreign investors in the Indian external sector policies is strong. India’s policies have also been described by the IMF as being "comparable to the global best practices" in a recent study of 20 select industrial and developing countries. Interestingly, a leading global news agency, in an international journal, has recently described India’s currency model as being "ideal" for Asia. India is now one of the very few developing countries which has set up its own clearing house for dollar-rupee transaction with the concurrence of the Federal Reserve System, New York.

26. In the last few months, however, when the dollar has been depreciating against major currencies, and rupee has been appreciating against the dollar (albeit slowly), a number of suggestions have been made by experts and others calling for a shift in RBI’s exchange rate policies. There are, in the main, three alternative approaches that have been suggested for consideration:

  • One view advanced by several distinguished economists, including Prof. Kenneth Rogoff of IMF during his recent visit to India, is that rupee should be allowed to appreciate freely in line with market trends. According to this view, there is no strong case for RBI’s further intervention as reserves are already very high. RBI’s purchases create substantial additional domestic liquidity, which may be destabilising in the long run. There is also no evidence, in their opinion, that unconstrained appreciation or volatility would affect growth prospects or lead to any other macro-economic problem.

  • An exactly opposite view, which, among others, has been recently articulated by an important all-India industry association, is that RBI should intervene more aggressively in the market to further reduce the degree of appreciation. The main argument in favour of this view is that India must maintain its global "competitiveness", particularly in relation to China which has a fixed exchange rate with the dollar and whose currency has been depreciating along with it.

  • A third view, which has been recently put forward by a leading economic journal, among others, is that RBI should pursue what it has referred to as a policy of "calculated volatility". It has been argued that the present policy of controlled volatility has provided virtually risk-less gain to market participants since the rupee has been expected to appreciate substantially and continuously over the past few months. According to this view, in order to prevent excessive capital inflows during this period, RBI should have allowed the exchange rate to "overshoot" quickly the targeted exchange rate of, say, Rs.46.20 (or any other number) to, say, Rs.45.50. Thereafter, it should have allowed the rupee to depreciate slowly, but not necessarily smoothly, to the above targeted number over a period of next few months. In essence, this proposal is akin to a policy of (announced or unannounced) fixed exchange rate within a wider band.

27. The RBI welcomes the current debate. Reserves, at present, are certainly at a level which is more than enough to meet any foreseeable contingency. It is also clear that, in the present period, capital inflows and remittances have been strong, requiring continuous domestic liquidity management. In principle, therefore, it would be nice if an alternative viable exchange rate management system could be put in place which would avoid excessive build-up of reserves and domestic liquidity and, at the same time, maintain India’s external competitiveness with low inflation and low interest rates.

28. In theory, each of the above alternative approaches has some merit. However, it is not entirely clear that they can be put into practice without causing substantial instability or uncertainty and possible emergence of macro-economic problems which are worse than what they are trying to solve. An implicit assumption in two of the above alternatives is that there is a level at which, after initial fast appreciation, the exchange rate will either stabilise or turn around. A further implicit assumption is that the level (whatever it is) is either already known or will become known to the market as it is approached.

29. RBI’s past experience does not suggest that these assumptions are valid. It would be recalled that there have been periods when rupee exchange rates have been relatively more volatile and movements have been sharper. However, during periods of sharper appreciation, instead of inflows declining and demand for foreign currency rising, it was noticed that actual market behaviour was the opposite. The opposite was true during periods of sharp depreciation. Exchange rate expectations had their own momentum and were often self-fulfilling. There must, of course, be a level where these expectations will reverse. However, if that level, because of "momentum" trading in imperfect and thin markets happens to be significantly out of line with "fundamentals", considerable instability and substantial overvaluation (or under valuation) may result. Such an outcome may do more harm than good to continued confidence in a country’s exchange rate system.

30. The third suggestion to hold the rates at current levels, and not to allow it to appreciate any further, even if inflows are strong, is also likely to be unsustainable over any length of time. It virtually amounts to adopting a "fixed" or a near-fixed exchange rate system with a floor. Past experience suggests that this system can work well, as it did in East Asia prior to the crisis, when the economy is doing well and inflows are strong, but it comes under extreme pressure when there are unfavourable domestic or external developments. Abandonment of a system of "fixed" exchange rates (or a system with a known floor) then becomes unavoidable. Such a change, when it occurs under pressure, can result in considerable instability which is likely to be spread over a fairly long period. At the end of this process, the country then has no option but to revert to a more flexible exchange rate system.

31. It is by no means a mere coincidence that all countries affected by external crises in the 1990s had a fixed or near-fixed exchange rate systems. China at present is an exception to the rule in view of its persistent trade surpluses over a long period combined with very high levels of foreign direct investment. It is not certain how long into the future this situation will prevail. In any case, China’s special characteristics are difficult to replicate in other emerging markets with lower volume of trade and foreign investment.

32. The desirability of maintaining the-overall competitiveness of an economy can hardly be questioned. However, the long-run competitiveness of an economy needs to be measured in relation to a multiple currency basket, and in relation to major trading partners over a reasonably long period of time. Exchange rate fluctuations among major currencies are now an everyday fact of life, and it is important for all entities with foreign exchange exposures to resort to "hedging" with appropriate risk management of assets and liabilities.

33. On balance, the benefits of the suggested alternatives to the present system are not very clear. The present system is by no means an ideal one. However, like the old cliche about virtues of democracy, it is probably better in the long run than all the available alternatives. In view of behavioural and market complexities in this area, as well as multiple economic policy objectives, solutions which seem "ex ante" optimal may turn out to be disastrous "ex post" - after the event – as happened in Argentina recently and East Asia and Mexico some years ago.

34. Nevertheless, as I said a while ago, RBI welcomes the present debate. As a contribution to this debate, I have tried to deal with some relevant issues, and indicate our present views on them. These views are, of course, subject to change in the light of domestic and international experience and further academic insights.

35. We look forward to your deliberations and hope to benefit from them.

Address by Dr. Bimal Jalan at 14th National Assembly of Forex Association of India 14.8.2003

Back to Speeches  

Previous  |  Next